A framework that recommends (suggests) an item related to an item that has been viewed or purchased based on a viewing history or a purchase history of a user in an online shopping service is known. Further, a recommendation service that sets a related item of a certain item or a recommended item and presents those items to a user is also known.
Relating to such services, an item recommendation system that determines a recommended item in a complex manner from multiple points of view is disclosed in Patent Literature 1 below. In this system, a recommendation server performs a hybrid recommendation process that determines a recommended item in a complex manner from two points of view with use of an item correlation information storage means and an individual item correlation information storage means and then displays the recommended item in a recommended item section in an item page.